1. Field of the Invention
The present invention is the in the field of telecommunications including dynamic voice recognition and speech rendering, and pertains particularly to methods for dynamically converting VoiceXML into other compatible markup languages based on required modality.
2. Discussion of the State of the Art
In state-of-art contact centers processing of incoming calls from customers typically involves interactive voice response (IVR) processing and/or live voice conversation with an agent of the center. IVR interaction dialog prompts are typically provided by the IVR system as synthesized speech rendered from voice XML script using a text-to-speech engine. Responses from customers are typically recognized by the speech engine and rendered to text for contact center application. Agent scripts are built using graphics user interface (GUI)-based tools. Therefore, both types of dialogs are created separately using different languages markup and tools.
IVR scripts and agent scripts dialogue types have several features in common. Both dialog types are directed to customers of the contact center. Both dialog types use contact center (CC) data. Both dialog types seek to provide information to or extract information from a customer. In a case where the IVR system is very “smart” a customer may not distinguish between synthesized speech and human speech because natural language technology is used. In natural language technology, the IVR recognizes natural customer speech and responds accordingly using dynamically created IVR scripting.
It has occurred to the inventors that it would be desirable to be able to dynamically switch dialog on the contact center side from dynamically created IVR scripting to agent scripting and back. However, because of differences in language construction, agent scripting cannot simply be substituted for IVR scripting. Some attempts have been made to develop abstract dialog management languages that have different presentation transformation capabilities, but a problem with this approach is in creating required abstraction in the language and managing complexity of the language.
What is clearly needed is a system and methods for dynamically converting voice XML scripting into other compatible markup language scripting based on required modality at the time of conversion.